检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Li, Kewei Kong, Yanwen Xu, Yiping Su, Jianlin Huang, Lan Zhang, Ruochi Zhou, Fengfeng College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Jilin Changchun130012 China Moonshot AI Ltd. Beijing100086 China

Since the emergence of research on improving the length extrapolation capabilities of large models in 2021, some studies have made modifications to the scaling factor in the scaled dot-product attention mechanism as part of their proposed methods without rigorous theoretical justifications. To fill this gap, we propose two new scaled temperatures based on entropy invariance to enhance length extrapolation. First, a training-free method InfoScale is designed for dot-product attention, and preserves focus on original tokens during length extrapolation by ensuring consistent entropy. Second, we theoretically analyze the impact of scaling (CosScale) on cosine attention. Experimental data demonstrates that combining InfoScale and CosScale achieves state-of-the-art performance on the GAU-α model with a context window extended to 64 times the training length, and outperforms seven existing methods. Our analysis reveals that significantly increasing CosScale approximates the Windowed Attention, and highlights the significance of attention score dilution as a key challenge in long-range context handling. The code and data are available at https://***/HT-NEKO/InfoScale. © 2025, CC BY-NC-ND.

关键词： Data handling

来源：评论

学校读者我要写书评

暂无评论

Multimodal Fusion for Android Malware Detection Based on Large Pre-trained Models

引用

IEEE Transactions on Software engineering 2025年第05期51卷 1569-1590页

作者： Li, Xun Liu, Lei Liu, Yuzhou Zhao, Yu Zhang, Peng Liu, Huaxiao Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China College of Computer Science University of Cincinnati OH United States

Malware detection is a critical issue in software engineering as it directly threatens user information security. Existing approaches often focus on individual modality (either source code or binary code) for the detection, but it ignores to effectively exploit the complementary information between them. This limits the detection performance, especially in complex and evasive malware scenarios. In this paper, we take Android applications written in Java as objects, and provide a novel fine-grained multimodal fusion method with large pre-trained models to combine the features from source and binary codes for the malware detection. For the source code modality, we employ the graphical user interface (GUI) as a framework to segment the source code into snippets, and use a pre-trained programming language model to extract feature representations. For the binary code modality, we convert binary code into grayscale images and fine-tune a pre-trained vision model to extract features indirectly. We then implement cross-modal attention and devise a contrastive loss to align features across modalities, supplementing this with supervised classification loss to refine the multimodal fusion process specifically for malware detection. Our experiments, conducted using the Data-MD and Data-MC benchmarks, demonstrate that our approach achieves a precision of 0.977 and a recall of 0.984 in detecting malware. This underscores the advantages of using large pre-trained models for feature representation and the fusion of information across different modalities for effective malware detection. © 1976-2012 IEEE.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Dual-Network Cross-Learning for Metabolite-Disease Association Prediction

IEEE Transactions on Computational Biology and Bioinformatic...

引用

IEEE Transactions on computational Biology and Bioinformatics 2025年第2期22卷 545-556页

作者： Yanxin Chen Qiao Ning Yitong Zhang Hui Li Shikai Guo Department of Information Science and Technology Dalian Maritime University Dalian China School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China Dalian Key Laboratory of Artificial Intelligence Dalian China

In recent years, increasing evidence has demonstrated a close association between metabolites and various complex human diseases, providing valuable insights for disease diagnosis, treatment, and prevention. Although deep learning-based approaches have achieved certain success in predicting metabolic disease associations, challenges remain in enriching graph information and effectively integrating metabolic and disease features. To address these issues, this paper proposes a model named DCMDA, which extracts deep features of both metabolites and diseases using Dual-network Cross-learning for Metabolite- Disease Association prediction. DCMDA consists of three parts. The data processing module integrates similarity networks with association networks to construct a heterogeneous network. The feature extraction module extracts features from the metabolite-disease association network based on the non-negative matrix factorization method and from the heterogeneous network using graph autoencoder techniques. The feature fusion module combines the association matrix feature with the heterogeneous network feature through a Cross-Attention mechanism, thereby obtaining deep representations of metabolites and diseases. These features are then used to train the model to predict association scores between metabolites and diseases. Experimental results demonstrate that in 5-fold cross-validation, DCMDA achieves an area under the receiver operating characteristic curve (AUC) of 97.8% and an area under the precision-recall curve (AUPR) of 97.9%, outperforming state-of-the-art prediction methods.

关键词： Diseases Feature extraction Data mining Semantics Deep learning Kernel Information entropy Prediction algorithms Noise Nearest neighbor methods

来源：评论

学校读者我要写书评

暂无评论

Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning

arXiv

引用

arXiv 2025年

作者： Liu, Yonghao Li, Mengyu Pang, Wei Giunchiglia, Fausto Huang, Lan Feng, Xiaoyue Guan, Renchu Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education College of Computer Science and Technology Jilin University China Mathematical and Computer Sciences Heriot-Watt University United Kingdom University of Trento Italy

Short text classification, as a research subtopic in natural language processing, is more challenging due to its semantic sparsity and insufficient labeled samples in practical scenarios. We propose a novel model named MI-DELIGHT for short text classification in this work. Specifically, it first performs multi-source information (i.e., statistical information, linguistic information, and factual information) exploration to alleviate the sparsity issues. Then, the graph learning approach is adopted to learn the representation of short texts, which are presented in graph forms. Moreover, we introduce a dual-level (i.e., instance-level and cluster-level) contrastive learning auxiliary task to effectively capture different-grained contrastive information within massive unlabeled data. Meanwhile, previous models merely perform the main task and auxiliary tasks in parallel, without considering the relationship among tasks. Therefore, we introduce a hierarchical architecture to explicitly model the correlations between tasks. We conduct extensive experiments across various benchmark datasets, demonstrating that MI-DELIGHT significantly surpasses previous competitive models. It even outperforms popular large language models on several datasets. Copyright © 2025, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Joint Mobile Edge Computing and Data Collection via Matching-Enabled Deep Reinforcement Learning

引用

IEEE Internet of Things Journal 2025年

作者： Wang, Boxiong Kang, Hui Li, Jiahui Sun, Geng Sun, Zemin Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China Nanyang Technological University College of Computing and Data Science 639798 Singapore

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) and data collection (DC) have been popular research issues. Different from existing works that consider MEC and DC scenarios separately, this paper investigates a multi-UAV-assisted joint MEC-DC system. Specifically, we formulate a joint optimization problem to minimize the MEC latency and maximize the collected data volume. This problem can be classified as a non-convex mixed integer programming problem that exhibits long-term optimization and dynamics. Thus, we propose a deep reinforcement learning-based approach that jointly optimizes the UAV movement, user transmit power, and user association in real time to solve the problem efficiently. Specifically, we reformulate the optimization problem into an action space-reduced Markov decision process (MDP) and optimize the user association by using a two-phase matching-based association (TMA) strategy. Subsequently, we propose a soft actor-critic (SAC)-based approach that integrates the proposed TMA strategy (SAC-TMA) to solve the formulated joint optimization problem collaboratively. Simulation results demonstrate that the proposed SAC-TMA is able to coordinate the two subsystems and can effectively reduce the system latency and improve the data collection volume compared with other benchmark algorithms. © 2025 IEEE. All rights reserved,

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE Internet of Things Journal 2025年

作者： Li, Yangning Kang, Hui Li, Jiahui Sun, Geng Sun, Zemin Wang, Jiacheng Zhao, Changyuan Niyato, Dusit Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun130012 China Nanyang Technological University College of Computing and Data Science 639798 Singapore

An expansion of Internet of Things (IoTs) has led to significant challenges in wireless data harvesting, dissemination, and energy management due to the massive volumes of data generated by IoT devices. These challenges are exacerbated by data redundancy arising from spatial and temporal correlations. To address these issues, this paper proposes a novel data-driven collaborative beamforming (CB)-based communication framework for IoT networks. Specifically, the framework integrates CB with an overlap-based multi-hop routing protocol (OMRP) to enhance data transmission efficiency while mitigating energy consumption and addressing hot spot issues in remotely deployed IoT networks. Based on the data aggregation to a specific node by OMRP, we formulate a node selection problem for the CB stage, with the objective of optimizing uplink transmission energy consumption. Given the complexity of the problem, we introduce a softmax-based proximal policy optimization with long-short-term memory (SoftPPO-LSTM) algorithm to intelligently select CB nodes for improving transmission efficiency. Simulation results show that the proposed OMRP improves network lifetime by 17% compared to benchmark routing protocols, while the SoftPPO-LSTM method for CB node selection achieves an 8.3% increase in throughput over benchmark algorithms. The results also reveal that the combined OMRP with the SoftPPO-LSTM method effectively mitigates hot spot problems and offers superior performance compared to traditional strategies. © 2014 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Prompt-guided orthogonal multimodal fusion for cancer survival prediction

引用

Information Sciences 2025年 715卷

作者： Huang, Lan Guo, Shuyu Bai, Tian Zhao, Ruihong Tao, Ke College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China Department of Gastroenterology The First Hospital of Jilin University Changchun130000 China

Cancer survival prediction can assist clinicians in developing personalized treatment plans for patients. Comprehensive cancer diagnosis and treatment require integrating macroscopic and microscopic imaging. However, significant discrepancies in the spatial resolution and anatomical scale between imaging modalities hinder existing multimodal fusion methods from effectively learning correlated semantic features with limited datasets. In this work, we introduce a prompt-guided orthogonal multimodal fusion strategy (POMF) for fusing multimodal medical images across anatomical scales. POMF utilizes modality-specific prompts to fine-tune pretrained models, facilitating bias adaptation to medical imaging features while ensuring more computationally efficient training. A modality consistency-discrepancy prototype is designed as the modality-inherent prompt in POMF, disentangling the multimodal features and bridging the potential correlations across the orthogonal modalities. POMF is validated on a glioma survival prediction task using paired radiology and pathology images. The experiment results suggest that POMF achieves superior C-index with existing full-tuning and prompt-tuning methods. Additionally, the ablation studies demonstrate that POMF is adaptable to various architectures of pretrained encoders and multiple multimodal fusion strategies on cross-scale medical images. © 2025 Elsevier Inc.

关键词： Mammography

来源：评论

学校读者我要写书评

暂无评论

Robustifying vision transformer for image forgery localization with multi-exit architectures

引用

Pattern Recognition 2025年 164卷

作者： Shi, Zenan Chen, Haipeng Zhang, Dong College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology 999077 Hong Kong

The proliferation of image manipulation tools has led to an increase in the number of manipulated images being disseminated online, posing risks like the propagation of fake news and telecom fraud. Thus, there is an increasing demand for precise, generic, and robust methods for detecting and locating manipulated images. In this paper, we propose a simple and clean model, named MEAFormer, for image forgery localization that does not heavily rely on pre-trained models. MEAFormer comprises three main components: an encoder network, a neck network, and a decoder network. Specifically, the transformer-based encoder network extracts hierarchical feature representations from the input image, providing rich contextual information in each layer. The neck network, incorporating our proposed cross-layer feature aggregation (CFA), aggregates these hierarchical features. To achieve better spatial feature co-occurrence, instead of using noise or edge artifacts, we introduce a multi-scale graph reasoning (MGR) module within the decoder network via bipartite graphs over the encoder and decoder features in a multi-scale fashion. The cross-level enhancement (CLE) further performs adjacent-level feature fusion to amplify the regions of interest in aggregated manipulation features. Finally, the multi-exit architecture (MEA) guides the model to learn fine-grained features and segment out the manipulated region. Extensive experiments across diverse and challenging datasets conclusively establish the superiority of MEAFormer over existing state-of-the-art methods, excelling in accuracy, generalization, and robustness. © 2025 Elsevier Ltd

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-Aware Global-View Reconstruction for Multi-View Multi-Label Feature Selection

arXiv

引用

arXiv 2025年

作者： Hao, Pingting Liu, Kunpeng Gao, Wanfu College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China Department of Computer Science Portland State University PortlandOR97201 United States

In recent years, multi-view multi-label learning (MVML) has gained popularity due to its close resemblance to real-world scenarios. However, the challenge of selecting informative features to ensure both performance and efficiency remains a significant question in MVML. Existing methods often extract information separately from the consistency part and the complementary part, which may result in noise due to unclear segmentation. In this paper, we propose a unified model constructed from the perspective of global-view reconstruction. Additionally, while feature selection methods can discern the importance of features, they typically overlook the uncertainty of samples, which is prevalent in realistic scenarios. To address this, we incorporate the perception of sample uncertainty during the reconstruction process to enhance trustworthiness. Thus, the global-view is reconstructed through the graph structure between samples, sample confidence, and the view relationship. The accurate mapping is established between the reconstructed view and the label matrix. Experimental results demonstrate the superior performance of our method on multi-view datasets. Copyright © 2025, The Authors. All rights reserved.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Reconsidering Feature Structure Information and Latent Space Alignment in Partial Multi-label Feature Selection

arXiv

引用

arXiv 2025年

作者： Pan, Hanlin Liu, Kunpeng Gao, Wanfu College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China Department of Computer Science Portland State University PortlandOR97201 United States

The purpose of partial multi-label feature selection is to select the most representative feature subset, where the data comes from partial multi-label datasets that have label ambiguity issues. For label disambiguation, previous methods mainly focus on utilizing the information inside the labels and the relationship between the labels and features. However, the information existing in the feature space is rarely considered, especially in partial multi-label scenarios where the noises is considered to be concentrated in the label space while the feature information is correct. This paper proposes a method based on latent space alignment, which uses the information mined in feature space to disambiguate in latent space through the structural consistency between labels and features. In addition, previous methods overestimate the consistency of features and labels in the latent space after convergence. We comprehensively consider the similarity of latent space projections to feature space and label space, and propose new feature selection term. This method also significantly improves the positive label identification ability of the selected features. Comprehensive experiments demonstrate the superiority of the proposed method. Copyright © 2025, The Authors. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：